Internati
o
nal Journal
of App
lied Power E
n
gineering
(IJAPE)
V
o
l.
2, N
o
. 2
,
A
ugu
st
2013
, pp
. 53
~60
I
S
SN
: 225
2-8
7
9
2
53
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJAPE
Optimal Placem
ent of D-S
T
ATCOM Usin
g Hybrid Gen
e
tic and
Ant Col
o
ny Algorithm
to Losses Reduction
Askar B
a
gher
inasab
1
, Mah
m
oud
z
a
dehb
agheri
1
, Sa
iful
niza
m Abdul
Kha
lid
1
,
Maji
d G
a
nd
omk
a
r
2
,
Naz
i
ha Ahm
a
d
Az
li
1
1
Faculty
of Electrical Engin
eerin
g, Universiti Tek
nologi Malay
s
ia, 81310 Skudai, J
ohor, Malay
s
ia
2
DEP
.
Ele
c
tri
c
a
l
Engine
ering.
Is
l
a
m
i
c Az
ad Univ
ers
i
t
y
,
S
a
veh
bra
n
ch,s
aveh
, I
r
an
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Dec 5, 2012
Rev
i
sed
Feb
21
, 20
13
Accepte
d
Mar 7, 2013
In this work,
a
m
odern algorith
m
b
y
h
y
brid g
e
n
e
ti
c algor
ithm
a
nd ant
colo
n
y
algorithm
is de
signed to p
l
a
c
e
m
ent and
then
sim
u
lated
to de
term
ine
th
e
am
ount of reacti
v
e power b
y
D-STATCOM. Also this m
e
thod wi
ll be able t
o
m
i
nim
i
ze the power s
y
s
t
em
los
s
es
that cont
ain power los
s
in trans
m
is
s
i
on
lines. F
u
rtherm
o
r
e, in this design
a IEEE 30-bus m
odel depict
ed and three D-
STATCOM are loc
a
ted
in
this s
y
stem
accord
ing t
o
Econom
ic
Cons
iderations
. The optim
al pla
cem
ent of each
D-S
T
ATCOM
i
s
com
puted
b
y
the
ant
co
lo
n
y
algor
ithm
.
I
n
order
to op
ti
m
i
ze pl
ac
em
ent
for e
ach
D-
STATCOM, two groups of ant are sele
cted
, which respectiv
ely
located in
near nest and far
from the nest. More
over, for ever
y
outpu
t simulation of D-
S
T
ATCOM
that is
us
ed to produce or abs
o
rb of reac
tive power
, a gene
ti
c
algorithm
to
m
i
n
i
m
i
zing th
e
tota
l
network losses i
s
appli
e
d. Fin
a
ll
y,
th
e resul
t
of this simulation shows net losses re
duction abo
u
t 150% that it
verifies the
new algor
ithm
p
e
rform
ance
.
Keyword:
An
t co
lon
y
algo
rith
m
D-S
T
ATC
O
M
Gen
e
tic al
g
o
rith
m
Op
tim
izat
io
n
Reactive powe
r
Copyright ©
201
3 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Saifu
l
n
i
zam
Ab
du
l Kh
alid,
Facu
lty of Electri
cal Engineering,
Un
i
v
ersiti Tekn
o
l
o
g
i
Malaysia,
8
131
0 Sk
ud
ai,
Jo
hor
, Malaysia.
Em
a
il: n
i
za
m@fk
e.u
t
m
.
my
1.
INTRODUCTION
Recently, Im
provi
ng of
powe
r
quality has
been c
o
nsidere
d
for
powe
r
distribution c
o
m
p
anies a
nd
b
o
t
h
low and
med
i
u
m
v
o
ltage co
st
u
m
ers [1]-[4
].
Th
ere are m
a
n
y
reaso
n
s fo
r m
o
re atten
tio
n to
po
wer qu
ality
of p
o
w
er di
st
ri
but
i
o
n com
p
an
i
e
s such as con
n
ect
net
w
or
ks t
oget
h
e
r
t
o
f
o
r
m
l
a
rger net
w
or
ks beca
use o
f
faul
t
ele
m
en
t in
th
e n
e
twork, in
creasin
g h
a
rm
o
n
ics in
p
o
we
r s
y
ste
m
s, customers'
increasing a
w
are
n
ess
of powe
r
q
u
a
lity issu
es,
in
creased
sen
s
i
tiv
ity o
f
electri
cal d
e
v
i
ces ag
ain
s
t d
i
sturb
a
n
c
es of
d
i
stribu
tio
n n
e
t
w
ork
s
[5]-[7
]
.
Because of the
rise of
unbala
nced l
o
adi
ng on
each pha
se, biased faults
al
ways take plac
e in the distribution
sy
st
em
whi
c
h
begi
nni
ng
s u
n
s
t
abl
e
v
o
l
t
a
ge a
nd
cu
rre
nt
wi
t
h
n
e
gat
i
v
e c
o
m
ponent
[
8
]
.
The
di
st
ri
b
u
t
i
o
n sect
o
r
as t
h
e m
a
i
n
l
i
nk
bet
w
een t
h
e pe
opl
e a
n
d t
h
e
po
wer
i
n
du
st
ry
rol
e
m
o
re
eval
uat
i
o
n a
n
d
ju
d
g
m
e
nt
t
h
an
ot
he
r
p
o
wer
p
a
rts an
d th
at's wh
y
th
e in
creasing qu
ality o
f
th
e electricity d
i
strib
u
tion
is essen
tial.
In add
itio
n
,
determ
ine the
optim
u
m
capacitor
placem
en
t in the
distribution system
is
us
e
d
t
o
m
i
nimize the energy
losses
wi
t
h
im
pro
v
i
n
g t
h
e v
o
l
t
a
ge p
r
o
f
i
l
e
of t
h
e sy
st
em
and t
h
en
enha
ncem
ent
of t
h
e p
o
we
r fac
t
ors o
f
a di
st
ri
but
i
o
n
syste
m
[9
]-[13]. Man
y
p
o
t
en
tial ap
p
licatio
ns su
ch
as
h
e
uristics an
d lin
ear
n
on-lin
ear op
timizatio
n
tech
niq
u
e
s
have
bee
n
e
x
pl
ore
d
t
o
s
o
l
v
e t
h
e
po
we
r
qual
i
t
y
pr
obl
em
[14
]
-[1
7]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-87
92
I
J
APE V
o
l.
2, N
o
. 2
,
Au
gu
st 2
013
:
53
–
6
0
54
2.
PRI
NCI
PLE OF D-ST
ATC
O
M
The D
-
ST
AT
C
O
M
has bee
n
ap
pl
i
e
d as a
fav
o
u
r
a
b
l
e
d
e
v
i
ce to
prov
ide an
i
m
p
o
r
tan
t
ro
le in
the
di
st
ri
b
u
t
i
on
sy
st
em
such as
vol
t
a
ge
sag m
i
t
i
g
at
i
on,
vol
t
a
ge st
abi
l
i
zat
i
on, fl
i
c
ker s
u
pp
ressi
o
n
,
po
we
r
fact
o
r
cor
r
ect
i
o
n
,
an
d
harm
oni
c con
t
rol
[1
8]
-
[
2
0
]
.
It
i
s
not
abl
e
t
h
at
D-ST
ATC
O
M
i
s
one of t
h
e im
port
a
nt
de
vi
ces
th
at are ab
le t
o
so
lv
e th
e
p
o
wer qu
ality p
r
o
b
l
em
s at
th
e
d
i
stribu
tio
n
n
e
twork
[21
]
-[2
2
]. D-STATCOM h
a
s
b
een used to
so
lv
e th
e
un
b
a
lan
ced
fau
lts in
th
e sy
stem
as a certain c
ont
roller
[22]. D-ST
ATCOM is a
n
u
n
b
i
ased
three-ph
ase
vo
ltag
e
o
r
curren
t
throu
g
h
an
ab
ility sh
un
t
d
e
v
i
ce so
th
at can con
t
ro
l t
h
e m
a
g
n
itu
d
e
and
the phase
angl
e [21].
Dist
r
i
butio
n
Static synch
r
on
ou
s co
m
p
en
sator
(D
-ST
A
TC
OM
) i
n
cl
ude
s a
vol
t
a
ge
sou
r
ce
inve
rter like a cont
roller, a
DC ener
gy
stora
g
e an
d Gate T
u
r
n
o
ff
(GT
O
)
thristor which
causes a bala
nced set
of c
u
rre
nt
o
r
t
h
ree
p
h
ases si
nus
oi
dal
v
o
l
t
a
ge at
t
h
e
ba
si
c fre
quency.
T
h
ere is a
n
e
ffi
cient control of both
act
i
v
e an
d
rea
c
t
i
v
e p
o
we
r
w
h
i
c
h
use
as c
o
nnect
i
n
g
de
vi
c
e
bet
w
ee
n t
h
e
D-S
T
ATC
O
M
an
d t
h
e
AC
s
y
st
em
.
Abs
o
rb
o
r
pr
o
d
u
ce c
ont
rol
l
a
bl
e act
i
v
e a
n
d
re
act
i
v
e p
o
we
r c
a
n
be a
ppl
i
e
d
b
y
D-S
T
ATC
O
M
co
nst
r
uct
i
o
n
[
2
]
.
Gene
rally, the core of a D-STATC
O
M made of a
t
h
re
e-p
h
ase i
n
vert
er w
h
i
c
h o
n
one si
de i
s
connected to t
h
e
network th
ro
ugh
th
e t
r
ansfo
r
m
e
r and
from
th
e o
t
h
e
r side end
to
a cap
acito
r wh
ich is its DC
po
we
r. I
n
ad
d
i
t
i
on, t
h
e i
n
pu
t
si
gnal
s
i
n
cl
u
d
e v
o
l
t
a
ge o
f
bus (
V
), o
u
t
put
cu
rre
nt
o
f
con
v
ert
o
r (
I
)
and a
refe
rence
voltage (dc
)
. T
h
e
real power is determined
by
refe
rence
v
o
l
t
a
ge w
h
i
c
h i
s
a
b
so
r
b
ed
by
t
h
e AC
syste
m
to
p
r
ovid
e
its in
tern
al
lo
sses.
3.
BASIC CONCEPTS OF
G
E
NETIC AL
GOR
ITHM
In
this pape
r Genetic Algori
thm
(GA)
is used
t
o
Placem
e
n
t and
determ
ine the
D-ST
ATCOM as
a
n
optim
al syste
m
[23].
Als
o
,
GA is a
n
e
ffecti
v
e m
e
thod in
bulky a
n
d e
x
te
nded places
that
has
code
d
variables
so
th
at led
t
o
th
e op
tim
al so
l
u
tio
n
[2
4
]
. The ad
v
a
n
t
age of co
d
e
d
v
a
riab
les is th
at th
e co
d
e
is t
h
e ab
il
ity to
trans
f
orm
a continuous s
p
ace
to a di
screte s
p
ace [25].
W
e
use the
GA m
e
th
od optimization i
n
population or a
set
of
p
o
i
n
t
s
i
n
a cert
a
i
n
m
o
m
e
nt
w
h
i
l
e
t
h
e
o
l
d m
e
t
hod
opt
i
m
i
zed has
bee
n
a
ppl
i
e
d
f
o
r
o
n
l
y
p
o
i
n
t
.
Thi
s
m
eans
that the large
num
ber
of projects ca
n be
proces
sed at
a
sam
e
ti
m
e
b
y
GA and
also
i
t
is n
o
t
ab
le that, GA
m
e
t
hod i
s
base
d
on
di
rect
ed
r
a
nd
om
ness.
In
or
der
t
o
use
o
f
G
A
m
a
ny
con
cept
s
s
u
ch
as
d
e
fi
ne
d t
h
e
o
b
j
e
c
t
i
v
e
function or cost function, de
finiti
on a
nd im
ple
m
entation of genetic sp
ace and de
finition and
im
ple
m
entation
of
G
A
ope
rat
o
r
s
.
4.
BASIC CONCEPTS OF ANT-CO
LONY OPTIMIZ
A
TION
(ACA)
In
recent years, extensive re
sear
ch on
optim
ization
m
e
thods has bee
n
use
d
for
s
o
lvi
ng dynam
i
c
pr
o
b
l
e
m
s
i
n
t
h
e fi
el
d
of
e
ngi
neeri
n
g
an
d
b
u
si
ness
as
wel
l
as n
u
m
e
rous
o
p
t
i
m
i
zat
i
on m
e
t
hods
ha
ve
bee
n
con
s
i
d
ere
d
[2
6
]
. The e
vol
ut
i
ona
ry
m
e
t
hods
t
h
at
are wel
l
-
kn
o
w
n
beca
u
s
e t
h
ei
r
uni
que
pr
o
p
ert
i
e
s are
m
o
re
con
s
i
d
ere
d
[
2
6
]
-[2
7]
. T
h
e a
n
t
col
o
ny
opt
i
m
izat
i
on
(AC
A
) i
s
o
n
e
of t
h
e
op
t
i
m
i
zat
i
on m
e
tho
d
s t
h
at
ha
ve
bee
n
i
nvest
i
g
at
e
d
i
n
evol
ut
i
ona
ry
cl
assi
fi
cat
i
on [
26]
-
[
28]
. T
h
i
s
opt
i
m
i
z
at
i
on t
echni
que
has
been i
n
spi
r
ed
by
t
h
e
b
e
h
a
v
i
our of
real an
ts fo
r findin
g
t
h
e m
eals b
y
o
p
tim
al p
e
rfo
r
m
a
n
ce as
h
a
s th
e ex
cep
tion
a
l ab
ility to
so
lve th
e
wel
l
-
k
n
o
w
n
o
p
t
im
i
zat
i
on o
f
e
ngi
neeri
n
g
an
d
b
u
si
nes
s
pr
ob
l
e
m
s
[26]
.
Wh
e
n
a
n
a
n
t
rem
oved
f
r
om
i
t
s
n
e
st
t
o
reach food, lea
v
es a
trace i
n
his path that is
a
chem
ical subs
tance called
pherom
one [26],[29].
Conse
que
ntly,
th
e o
t
h
e
r an
ts g
u
i
d
e
to
locate fo
od
with
scen
t an
d
fo
llow th
e p
a
th
mark
ed
ou
t, so
far th
e an
ts leav
ing
p
h
e
ro
m
o
n
e
in
t
h
eir
p
a
th to
add
to its con
c
en
tratio
n [2
6
]
.
Fi
g.
1. t
h
e t
e
st
pat
h
s
o
f
a
n
t
s
If ex
ist sev
e
ral p
a
th
s
with
d
i
ssi
m
ilar d
i
stan
ces b
e
tween
th
e n
e
st and
th
e
fo
od
o
f
th
e an
t
s
, sho
r
test
path will be opti
mized because
m
o
re phe
rom
one
of differe
n
t ants re
m
a
ined
behind the shortest path as shown
in fi
gu
re
(1
).
In
this sc
hem
a
tic, at
first, two
p
a
th
s th
at co
n
s
ist
of
ACB and
ADB sel
ect b
y
an
ts,
After a
n
u
m
b
e
r
of an
ts in
t
h
e sho
r
ter
p
a
th ACB i
n
creased
wh
ile
the num
b
er
of a
n
ts
decrease
s
i
n
t
h
e
prolonge
d
path
(ADB), acco
r
din
g
l
y, all th
e an
ts m
o
v
e
in th
e sho
r
test
way.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
APE
I
S
SN
:
225
2-8
7
9
2
Opt
i
m
al
Pl
ace
m
ent
of
D-
STA
T
C
O
M
Usi
n
g
Hybri
d
Ge
net
i
c
a
n
d
A
n
t
C
o
l
o
ny Al
gori
t
h
m (
A
skar
Ba
g
h
eri
nas
a
b
)
55
5.
POWER SYSTEM SIMULATION
Si
m
u
latio
n
resu
lts o
n
a 30
bu
s of th
e IEEE are in
v
e
stigated that is one of the fam
ous system
s
of
p
o
wer qu
ality. After sim
u
lati
o
n
o
f
3
0
bu
s of IEEE, Pl
acemen
t
Op
timize
r
o
f
D-STATC
O
M is d
e
sign
ed
and
th
en
d
e
term
in
e of D-STATC
O
M reactiv
e po
wer is calcu
lated
b
y
usi
n
g
a co
mb
ina
tion
o
f
G
A
a
n
d
AC
A.
The used power sys
t
em
One
IE
EE
30
-
bus
m
odel
use
d
s
o
t
h
at
5
n
u
m
ber of i
t
s
b
u
s
ha
ve a
ge
nerat
o
r
as s
h
o
w
n i
n
fi
g
u
re
(
3
).
Fi
g. 3.
IE
EE 3
0
-B
us
M
odel
Data of tran
sm
issio
n
lin
es is
on
e
requ
ired
info
rm
atio
n
in
sy
ste
m
si
m
u
latio
n
;
th
erefo
r
e, it
sh
ou
l
d
b
e
rem
i
n
d
e
d
th
at th
is system
h
a
s 4
1
transmissio
n
lin
es.
6.
PLACE
MEN
T
OF
D-
STA
T
CO
M I
N
NETWOR
K
Determ
ine
the require
d
num
ber of D-ST
AT
COM,
placem
ent and also t
h
e am
ount
of
re
active power
gene
rat
e
d
or a
b
so
r
b
ed
by
t
h
e D-S
T
ATC
O
M
i
s
very
si
g
n
i
f
i
cant
st
age
t
o
D-
STA
T
C
O
M
desi
g
n
i
n
g.
Due t
o
econom
i
c considerations espe
cially
in powe
r syste
m
s
three sim
i
lar D-ST
ATCOM is
ins
t
alled. Becaus
e
, the
syste
m
has 30 bus
es, so, 30 choices ar
e
for the location of
each D-ST
AT
COM.
Ge
neral
l
y, one of bus uses t
o
refe
rence
b
u
s,
and
t
h
e
r
ef
ore
,
di
ffe
re
nt
m
odes f
o
r
3
D
-
ST
A
T
C
O
M
i
n
st
al
l
e
d at
29
b
u
ses
i
s
as f
o
l
l
o
ws.
)!
1
(
!
)!
1
(
r
n
r
n
s
(
1
)
Wh
ere
(n) is total o
f
bu
s and
(r) is th
e
nu
m
b
er
o
f
D-ST
ATC
O
M.
Also, in ord
e
r to ob
tain
p
o
s
sib
l
e states(s)
for
th
ree
nu
m
b
er of in
stalled
D-STAT
C
O
M in t
h
e system
, we
can
write.
3654
)!
3
1
30
(
!
3
)!
1
30
(
s
(
2
)
Thu
s
,
36
54 non
-rep
etitiv
e m
o
d
e
is av
ailab
l
e fo
r t
h
e i
n
stallatio
n
o
f
D-STATCOM
in
t
h
e po
wer system
an
d
also
,
find
s an
op
ti
m
a
l p
o
i
n
t
i
n
all cases is
f
o
c
u
se
d
by
t
h
e c
o
m
b
i
n
at
i
on o
f
A
G
a
n
d
G
A
.
7.
GENETIC ALGORIT
HMS TO PL
ACEMENT
AND DETERMINATION
OF
D-STAT
COM
Due to im
port
a
nt designing
of
active loa
d
s
in each bus of D-ST
ATCOM,
the active and
reactive
po
we
r
of eac
h
sy
st
em
i
s
det
e
rm
i
n
ed as
fol
l
o
w
s.
p
=
[
4
0
.
00
00
-
2
.400
0
-7.60
0
0
-9
4.2000
0
-22
.
8
000
-
3
0
.
00
00
0
-
5
.80
0
0
0
-11
.
2
000
4
0
.000
0
-6
.2
00
0
-8
.2
000
-3
.5
000
-9
.00
0
0
-3
.20
0
0
]
-
9
.
5
0
0
0
-
2
.
2
0
0
0
-
1
7
.
5
0
0
0
4
0
.
0
0
0
0
4
0
.
0
0
0
0
-
8
.
7
0
0
0
0
-
3
.
5
0
0
0
4
0
.
0
0
0
0
0
-
2
.
4
0
0
0
-
10
.6
0
00]
(M
W)
Q
=
[
5
.300
0
-
1
.20
0
0
-
1
.600
0
-
1
9
.
00
00
0 -10
.
9
000
-3
0.000
0
0
-
2
.000
0
0
-7
.5000
2
5
.000
0
-1
.60
0
0
-2
.5
000
-
1
.8
000
-
5
.8
000
-
0
.900
0
-
3
.40
0
0
-
0
.700
0
-11
.
2
000
1
5
.0000
8
.
4
000
-6
.70
0
0
0
-2
.3
00
0
30
.0
000
0
-0
.9
000
-
1
.900
0
]
(
M
VA
R)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-87
92
I
J
APE V
o
l.
2, N
o
. 2
,
Au
gu
st 2
013
:
53
–
6
0
56
As can be see
n
, the val
u
es of active and re
active po
wer a
t
t
h
e refere
nce
bus (
b
u
s
1) i
s
uni
de
nt
i
f
i
e
d
so,
t
h
ese u
n
i
d
e
n
t
i
f
i
e
d p
o
we
rs can be
s
o
l
v
e
d
by
t
h
e Ne
wt
o
n
–R
ap
hs
on
m
e
tho
d
.
There
i
s
not
c
o
m
p
ensat
i
on
by
fi
rst
bus
i
n
t
h
e
ran
g
e
of
2
t
o
3
0
, s
o
,
c
o
st
f
u
nc
t
i
on ca
n
be
def
i
ned a
s
.
(
1
0
0
0
[
(
(
1
)
(
2
))
])
(
1
000
[
(
(
1
)
(
3
)
]))
(
1
000
[
(
(
2
)
(
3
)
]
1
1
000
1000
100
0
roun
d
x
x
r
ou
nd
x
x
r
o
und
x
x
ee
e
F
(
3
)
Wh
ere x
(1), x
(2)
an
d
x
(3) are
lo
cation
o
f
D-STATC
O
M
resp
ectiv
ely and
also, If th
e selected
lo
cation
will
be repeat
e
d
, F
1
=0.
The
following
equation is
use
d
to placem
ent
x
(4) a
n
d x (6)
in the
ra
nge
of
2 to 30.
1000
(
(
1
)
31
)
1000
(
(
2
)
31
)
1000
(
(
3
)
31
)
1
000
(
1
(
1
)
)
1000
(
1
(
2
)
)
1000
(
1
(
3
)
)
2
10
00
1000
100
0
100
0
1000
100
0
XX
X
x
x
x
ee
e
e
e
e
F
(4)
Whe
r
e
x (4), x
(5)
and
x (6), respectively, represe
n
t
th
e re
active powe
r generate
d or abs
o
rbe
d
by each
of the
D-S
T
ATC
O
M
.
Also
F
3
is applied
in
th
e range -50
M
W to +5
0
M
W
t
o
calc
u
late the am
ount of
reactive
powe
r
gene
rat
e
d
o
r
a
b
so
r
b
ed
by
eac
h
of
t
h
e
D-
ST
ATC
O
M
a
s
f
o
l
l
ows.
1
000
(
(
1
)
50
)
1
0
0
0
(
(
2
)
5
0
)
10
00
(
(
3
)
5
0
)
1
000
(
5
0
(
2
)
)
100
0
(
50
(
3
)
)
1
000
(
5
0
(
1
)
)
3
Xx
x
x
x
x
e
ee
e
e
e
F
)
5
(
After
determ
ining the am
ount and placem
ent of eac
h D-
STATC
O
M in its bus
by GA
and t
h
en loss of the
en
tire
n
e
two
r
k is calcu
lated
b
y
Newton- R
a
wson Meth
od will b
e
calcu
l
a
ted
,
sub
s
equ
e
n
tly, co
st
fu
n
c
tio
n
to
m
i
nim
i
ze i
s
de
fi
ne
d
by
t
h
e
ge
net
i
c
al
g
o
ri
t
h
m
,
so
,
we ca
n
wr
i
t
e
.
12
3
t
F
FF
F
The following values
are
selected to m
i
nim
i
ze F
t
b
y
th
e GA
.
Pop
u
l
ation
size = 40
, Mu
tation
fun
c
tion
=
Gau
ssian, M
u
tatio
n scale =
1
a
nd
Mu
tation
shrin
k
= 1
Th
e sim
u
latio
n resu
lts in
is
sho
w
n
as below.
Tabl
e 1. Si
m
u
lat
i
on resul
t
s
of
t
h
e GA
Reactive power
output
Nu
m
b
er
of bus
3138
.
25.
4
D-ST
ATCO
M.1
23.
534
8
5
D-ST
ATCO
M.2
32.
450
1
10
D-ST
ATCO
M.3
To
tal system
lo
sses
with
an
d withou
t the D-STAT
COM th
at is in
clu
d
e
d
the
p
o
wer lo
sses in to
tal
t
r
ansm
i
ssi
on l
i
n
es a
r
e s
h
ow
n
i
n
Ta
bl
e (
2
)
.
Tabl
e 2.
syste
m
losses
(M
W)
5.
1685
L
o
sses W
ithout
STATC
O
M
D-
3.
6441 (
M
W
)
Losses with
D-ST
ATCO
M
The
v
o
l
t
a
ge
ra
nge
o
f
ea
ch
b
u
s i
n
t
h
e
sy
st
e
m
wi
t
hout
D-
STATC
O
M
a
n
d
wi
t
h
D-
ST
A
T
C
O
M
i
s
s
h
o
w
n
i
n
fig
u
re (
2
)
a
s
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
APE
I
S
SN
:
225
2-8
7
9
2
Opt
i
m
al
Pl
ace
m
ent
of
D-
STA
T
C
O
M
Usi
n
g
Hybri
d
Ge
net
i
c
a
n
d
A
n
t
C
o
l
o
ny Al
gori
t
h
m (
A
skar
Ba
g
h
eri
nas
a
b
)
57
Fig
.
2
.
Pro
f
ile
vo
ltag
e
in 30-bus system
with
an
d withou
t D-STATC
O
M
As sh
own
in
figu
re
(2) ex
istin
g
o
f
three
D-STATC
O
M
in
th
e system n
o
t
on
ly redu
ces lo
sses,
bu
t also
i
m
p
r
ov
es th
e vo
ltag
e
p
r
o
f
ile an
d in
creasing
t
h
e
v
o
ltag
e
of all b
u
s
in
po
wer syste
m
.
8.
A HYB
R
ID ALGORIT
HM
WITH ACAF AND GA
TO DETERMINE THE BOT
H
VAL
UE AND
PLACE
MEN
T
OF
D-
STA
T
CO
M
So as t
o
determine the
Placement of three
D-ST
ATC
O
M
in
30bus
network,
ACA is
use
d
and for
determ
ining the am
ount
of
re
active powe
r generate
d or
a
b
s
o
rbe
d
by eac
h
D-S
T
ATC
O
M, GA is
utilized at all
stages.
Fi
g.
3. Fl
o
w
cha
r
t
o
f
pr
o
pose
d
m
e
t
hod
As s
h
ow
n i
n
th
e flo
w
c
h
art in
Figu
re
(3
),
firs
t, a
po
in
t is cho
s
en
rand
o
m
ly
b
e
tween
1
and 36
54
. Th
is
poi
nt
i
s
obt
ai
n
e
d of t
h
e e
qua
t
i
on (
1
) w
h
ere
t
h
i
s
poi
nt
i
s
t
h
e n
o
n
-rec
u
rri
ng l
o
cat
i
o
n o
f
D-ST
ATC
O
M
at
3
i
n
st
al
l
a
t
i
on b
u
s
bet
w
ee
n 2 t
o
30
of
b
u
s. T
h
e
ei
t
h
er o
p
t
i
m
u
m
val
u
es
of re
act
i
v
e po
we
r g
e
nerat
e
d o
r
a
b
sor
b
e
d
o
f
th
ree D-STATCOM is calcu
l
ated
b
y
GA By to re
d
u
c
e th
e to
tal syste
m
lo
sses.
Op
ti
m
i
zatio
n
b
y
GA is
applied to t
h
e
three
varia
b
les
as x (1),
x
(2) a
n
d x
(3) which these
va
ria
b
les re
present each of
the re
active
p
o
wer o
f
D-STATCOM resp
ectiv
ely. Th
e GA will b
e
stopp
ed
after 200
rep
e
titio
n
s
. Subsequ
e
n
t
to
th
is cycle
(200
rep
e
titio
n), t
h
e lowest lev
e
l of
n
e
two
r
k
lo
sses
will b
e
sto
r
ed
as
b
e
st an
swer.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-87
92
I
J
APE V
o
l.
2, N
o
. 2
,
Au
gu
st 2
013
:
53
–
6
0
58
In add
itio
n, a
lo
op
is
fo
rm
e
d
th
at
th
e
num
b
e
r (N
)
o
f
iteratio
n
s
is
1
0
0
0
.
Sub
s
eq
u
e
ntly, to
red
u
ce
l
o
sses,
AC
A i
s
em
pl
oy
ed by
sm
art
searchi
ng i
n
t
h
e best
poi
nt
of t
h
e w
hol
e
net
w
or
k t
o
i
n
st
al
l
t
h
ree
of
D-
STATC
O
M
.
T
w
o
ra
n
dom
poi
nt
s
near t
h
e
ne
st
an
d a
w
ay
t
h
e nest
i
s
sel
ect
ed t
o
o
b
t
a
i
n
C
o
st
fu
nct
i
o
n as
b
e
l
o
w.
3
F
Loss
Power
Total
F
T
(
3
)
In al
l
st
ages t
h
e al
gori
t
hm
wi
ll
be st
op
pe
d af
t
e
r 10
0
0
cy
cl
e and al
s
o
exi
s
t
2
00 i
t
e
rat
i
o
ns
of
t
h
e ge
net
i
c
algorithm
both nea
r
the
nest
and away
the
nest.
As, 400,000 iterations
ar
e total
reps
t
o
reach the
optim
a
l
so
lu
tion
.
Si
mul
a
ti
on re
sul
t
s
of
the
pr
op
osed
al
gori
t
hm
To
tal lo
sses of
d
i
stribu
tio
n lines are as two
m
o
d
e
s, on
e
with
three
D-STATCOM wit
h
tab
l
e (3
) cond
itio
n
s
,
Tabl
e.
3.
The r
e
sul
t
s
o
f
t
h
e
pr
op
ose
d
al
go
ri
t
h
m
Reactive power
production (
M
VA)
Nu
m
b
er
of bus
35.
982
8
4
D-ST
ATCO
M
1
45.
799
8
10
D-ST
ATCO
M
2
-
16.
8038
15
D-ST
ATCO
M
3
An
d t
h
e sec
o
n
d
,
wi
t
h
out
D
-
STATC
O
M
i
s
gi
ve
n i
n
t
a
bl
e
(
4
)
.
Tabl
e.
4.
Losse
s of system
6.
1685 (
M
W
)
L
o
sses W
ithout D-ST
AT
C
O
M
2.
7934 (
M
W
)
Losses W
i
t
D-S
T
A
T
COM Ac
cording Table
(3)
As sh
o
w
n i
n
Tabl
e (
4
) t
h
e l
o
ss rat
e
of a s
y
st
em
wi
t
h
t
h
ree D-S
T
ATC
O
M
i
s
dr
o
ppe
d t
o
4
5
.
2
84
9
.
Sin
ce th
e u
s
efu
l
life o
f
a p
o
wer system is
eq
u
a
l to
30
years, sign
ifican
t
l
y red
u
ce th
e co
st will b
e
p
r
ep
ared
in
pr
o
duct
i
o
n a
n
d
t
r
ansm
i
ssi
on of
po
we
r. Sy
st
em
vol
t
a
ge p
r
ofiles in
two
d
i
fferen
t m
o
d
e
s, with
and
with
ou
t
th
e
th
r
ee
D
-
STA
T
CO
M system
i
s
show
n in
f
i
gur
e (4)
.
Fig
.
4
.
System
v
o
ltag
e
pro
f
iles, wit
h
an
d wit
h
ou
t th
e three
D-STATC
O
M
As s
h
o
w
n i
n
f
i
gu
re (
4
)
,
usi
n
g D
-
ST
ATC
O
M
i
n
t
h
e net
w
or
k n
o
t
onl
y
r
e
duce
s
o
h
m
i
c
l
o
sses i
n
t
h
e
t
r
ansm
i
ssi
on s
y
st
em
, but
al
so
si
g
n
i
f
i
cant
l
y
i
m
proves t
h
e
v
o
l
t
a
ge
pr
ofi
l
e
s.
Com
p
ari
s
on
of t
h
e resul
t
s
of t
h
e tw
o
metho
ds pr
ovi
de
d (
g
enet
i
c
al
gori
t
hm
and t
h
e pr
op
ose
d
al
g
o
ri
thm
)
Th
e co
m
p
arison
o
f
th
e t
w
o meth
od
s is
shown
in tab
l
e
(5)
briefly as:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
0.
85
0.
9
0.
95
1
1.
05
1.
1
Bu
s N
u
m
b
e
r
V
o
l
t
age (
pu)
W
i
t
h
D
-
ST
AT
C
O
M
W
i
th
o
u
t D
-
S
T
A
T
C
O
M
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
APE
I
S
SN
:
225
2-8
7
9
2
Opt
i
m
al
Pl
ace
m
ent
of
D-
STA
T
C
O
M
Usi
n
g
Hybri
d
Ge
net
i
c
a
n
d
A
n
t
C
o
l
o
ny Al
gori
t
h
m (
A
skar
Ba
g
h
eri
nas
a
b
)
59
Tabl
e
5. C
o
m
p
ari
s
o
n
of
t
h
e
re
sul
t
s
o
f
ge
net
i
c
al
go
ri
t
h
m
and
t
h
e p
r
op
ose
d
al
go
ri
t
h
m
Fi
g.
5.
Vol
t
a
ge
pr
ofi
l
e
s at
t
h
re
e di
f
f
ere
n
t
m
e
tho
d
s
As sho
w
n in
figu
re (5
)Vo
ltag
e
p
r
ofiles in
t
h
ree
different states is i
nve
stig
ated
as, withou
t D-
STATC
O
M
,
with
D-
STA
T
COM
w
h
ich
it
s prop
erties is calcu
lated
b
y
th
e
g
e
n
e
tic al
g
o
rith
m
an
d
with
D-
STA
T
C
O
M
b
y
pr
opo
sed m
e
t
h
od
.
9.
CO
NCL
USI
O
N
Recently, i
m
provi
ng
of powe
r
quality has been c
ons
i
d
ere
d
for c
o
m
p
ensation
of
reactive
powe
r a
nd
h
a
rm
o
n
i
cs
b
e
cau
s
e to so
lv
e
th
e prob
lem
o
f
op
tim
u
m
reco
nfigu
r
ation
i
n
d
i
stribu
tio
n
syste
m
s, an
op
ti
m
a
l
m
a
nner
has
be
en nee
d
e
d
. T
h
i
s
pape
r p
r
ese
n
t
s
a ne
w a
p
p
r
oac
h
f
o
r o
p
t
i
m
al
m
a
nner o
f
di
st
ri
b
u
t
i
o
n
s
y
st
em
s
wh
ich
ACO is u
s
ed to d
e
termin
e
t
h
e
pl
ac
em
ent
o
f
t
h
ree
o
f
D
-
ST
ATC
O
M
i
n
30
b
u
s
net
w
or
k
an
d
GA
i
s
utilized for
determ
ining the
am
ount
of reactive power ge
nerate
d
or
abs
o
rbe
d
by
each D-ST
ATCOM.
In
stallatio
n
and
u
tilizatio
n
o
f
th
e D-STATCOM in
d
i
stri
b
u
tion
n
e
t
w
ork
s
lead
s to
esp
ecially sig
n
i
fican
t for
n
e
two
r
k
q
u
a
lities su
ch
as
reducin
g
o
f
oh
m
i
c l
o
sses i
n
tr
an
smissio
n
lin
es, imp
r
ov
e vo
ltag
e
p
r
o
f
iles an
d
syste
m
efficiency.
Finally,
m
a
intenance c
o
sts of the D-STAT
C
O
M
i
n
di
st
ri
b
u
t
i
on
net
w
o
r
ks
a
nd
p
o
we
r sy
st
em
s are
n
e
g
lig
i
b
le so
t
h
at th
e
en
er
g
y
s
a
v
i
ng
s
and
eco
n
o
m
izin
g
will b
e
sign
ifican
t.
REFERE
NC
ES
[1]
Ad
y
a
, A., e
t
al
. “
A
pplication o
f
D-STATCOM for is
olated s
y
stem
s, in Tenco
n
2004 - 2004
Ieee Reg
i
on 10
Conferenc
e
, Vol
s
a-D”,
Proceed
ings: Analog an
d Digital Tech
n
i
ques in Electrical Eng
i
neering
. Pp. C351-C354,
2004.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
252
-87
92
I
J
APE V
o
l.
2, N
o
. 2
,
Au
gu
st 2
013
:
53
–
6
0
60
[2]
Barnes, M., et al. “Power
quality improvement for wave en
erg
y
convert
ers using
a D-STATCOM with real en
er
g
y
s
t
orage”
,
2004 1st Internationa
l Conferenc
e
on Power Elec
tr
onics Systems and Applica
tions
Proceedings
, e
d
.
K.W.E. Ch
eng.
Pp. 72-77
, 2004
.
[3]
Cetin
, A.,
et
al.
“Reactiv
e power
compensation o
f
coal
convey
o
r
belt dr
ives b
y
u
s
ing d-statcoms”, in
Conf
er
enc
e
Record of th
e 2007 Ieee Industr
y Appli
c
ations
Conferenc
e
Fort
y-Second Ias An
nual Meet
ing
, V
o
l. 1-5. Pp. 173
1-
1740, 2007
.
[4]
Cai,
R.
, e
t
al.
“
C
ontrol of
D-STATCOM for v
o
ltag
e
dip
m
itig
ation
”
,
2005
International Con
f
erence on
Futur
e
Powe
r Sy
ste
m
s,
Pp. 126-131, 20
05.
[5]
Blazic, B. and I
.
Papic. “A new
mathematical model and control of
D-StatCom for operation under unbalanced
conditions”,
El
e
c
tric
Power
Syst
ems Research
, V
o
l/Issue: 72(3)
.
Pp. 279-287, 20
04.
[6]
Somsai, K.,
T. Kulworawanichpong, and
Ieee. “Modeli
ng, Simulation and
Contro
l of D-
STATCOM using
ATP
/
EMTP
”, in
2008 13th International Conf
erence on
Harmonics and Quality of Power
, Vol.
1 and 2. Pp. 377-
380. 2008
.
[7]
Niknam, T., H.Z. Mey
m
and
,
and M. Nay
e
ripo
ur. “A pr
actical algorithm for optimal operatio
n management of
distribution
network including
fu
el
cel
l power
pl
a
n
ts
”,
Ren
e
wable Energy
, Vol/Issu
e: 35(8)
. Pp. 169
6-1714, 2010
.
[8]
Noroozian, R. “A Performance
Comparison of D-STATCO
M and DC Distribution S
y
stem for
Unbalanced Lo
ad
Compensation”,
International Review
of
Electrica
l Eng
i
neering-Ir
ee
, Vol/Issue: 7(
2). Pp. 4194-420
7, 2012
.
[9]
Barukcic, M., S. Nikolovski, and
F. Jovi
c. “
H
y
b
r
i
d evolution
a
r
y
-h
euris
t
i
c
al
gori
t
h
m
for capacitor
banks alloc
a
tion
”
,
Journal of Electrical
Engine
erin
g-Elektrotechnicky Casopis
, Vol/Issue: 61(6). Pp.
332-340, 2010
.
[10]
S
ecui, D.C
., e
t
a
l
. “
A
n ACA Algorithm
for Optim
al Capaci
tor Banks
P
l
acem
ent
in P
o
wer Dis
t
ribution Networks
”
,
Studies
in Infor
m
atics and Con
t
rol
, Vol/Issue: 1
8
(4). Pp. 305-31
4, 2009
.
[11]
P
a
dm
anaban, K
.
P
.
and
G. P
r
ab
haharan
.
“
D
y
n
a
m
ic ana
l
y
s
is
on
optim
al p
l
a
c
e
m
ent of fixtu
r
i
ng el
em
ents
us
i
n
g
evolution
a
r
y
techniques”,
International Journal
of Production
Research
, Vol/Issue: 46(15)
. Pp. 4
177-4214, 2008
.
[12]
Su, C.T., C.F. C
h
ang, and J.P. Chiou.
“
O
ptim
al capa
c
itor pl
ac
e
m
ent in dis
t
ribution s
y
s
t
em
s
emplo
y
ing ant colo
n
y
s
earch algor
ithm
”
,
Electric
Power Components a
nd Systems
, Vol/Issue: 33(8). Pp.
931-946, 2005
.
[13]
Chang, C.-F. “Reconfigur
ation and Cap
aci
tor P
l
a
cem
ent for Los
s
Reduction of
Distribution S
y
stems by
Ant Colon
y
S
earch Algor
ith
m
”
,
Ieee Transa
c
tions on
Power
Systems
, Vol/Issue: 23(4)
. Pp. 17
47-1755, 2008
.
[14]
Atighehch
i
an, A
., M
.
Bijari,
and
H.
Tarkesh. “A novel h
y
brid
algorithm
for sch
e
duling steel-making continuous
casting
production”,
Computers &
Operations
Research
, Vol/Issue: 36(8)
. Pp. 24
50-2461, 2009
.
[15]
Sedki, A. and
D. Ouazar
. “H
y
b
r
i
d partic
le swarm optimization
an
d differ
e
ntial
ev
olution for
optimal design of
water
distribution s
y
ste
m
s”
,
Ad
vanced
Engineering
Info
rmatics
, Vol/Iss
ue: 26(3)
. Pp. 58
2-591, 2012
.
[16]
Saffar, A
., R
.
H
ooshmand, and
A. K
hodabakhsh
i
an. “A new fuzzy
optimal reco
nfiguration of d
i
stribution s
y
stems
for loss reductio
n and load balan
c
ing usi
ng ant c
o
lon
y
s
e
arch-b
as
ed algori
t
hm
”,
Applied Soft Computing
, Vol/Issu
e:
11(5). Pp. 4021-
4028, 2011
.
[17]
Valenzu
e
la
, C.,
et al
. “
A
2-level Meta
heuristic for the Set Covering Problem”,
International Jou
r
nal of Computers
Communications &
Control
, Vol/Issue: 7(2). Pp. 3
77-387, 2012
.
[18]
Mariun, N., e
t
al
. “
D
esign of a protot
ype D-stat
c
o
m
using
DSP controlle
r for volt
a
ge sag m
itigat
i
on’,
IEEE Powe
r
India Conference
, Vol. 1
and
2.
Pp. 727-732, 20
06.
[19]
Xi, Z., et al. “Improving Distributi
on S
y
s
t
em
P
e
rform
ance with
Integrat
ed S
T
ATCOM
and S
u
percap
ac
itor Ener
g
y
S
t
orage S
y
s
t
em
”
,
in
2008 I
e
ee
Po
wer Electronics
Specia
lists
Conference
, Vol. 1-1
0
. Pp. 1390-139
5, 2008
.
[20]
P
a
rkhideh, B
., e
t
al. “
I
ntegr
a
tion
of S
upercapa
c
ito
r with
S
T
ATCOM for Electr
i
c A
r
c F
u
rnace F
l
i
c
k
e
r Mitiga
tion”
, i
n
2008 Ieee
Power
Electronics
Specialists Con
f
eren
ce
, Vol. 1-10. Pp
. 2242-2247
, 20
08.
[21]
Coteli
, R., e
t
al. “
P
has
e
A
ngle
Control of Three Level In
verter
Based D-STATCOM Using Neuro-Fuzzy
Controller
”
,
Ad
vances in
Elec
trical and Comput
er Engin
eering
, V
o
l/Issue: 12(1)
.
Pp. 77-84, 2012.
[22]
Coteli, R
.,
et
al. “Three-level C
a
scad
ed
Inverter
Based D-STATCOM Using D
ecoupled
Indirect Current Con
t
rol”,
Iete Journal of
Research
, Vol/Issue: 57(3)
. Pp. 20
7-214, 2011
.
[23]
Afs
h
ar, M
.
H. “
L
arge s
c
al
e res
e
rv
oir oper
a
tion
b
y
C
onstrained
Part
icl
e
Swarm
Opti
m
i
zation
algor
it
hm
s”,
Journal o
f
Hydro-Environ
m
ent Research
,
Vol/Issue: 6(1)
.
Pp. 75-87, 2012.
[24]
Alm
e
der, C. an
d L. M
o
ench. “
M
etaheur
i
s
tics
f
o
r s
c
heduli
ng jo
bs with incom
p
atibl
e
fa
milies on para
llel batching
m
achines
”,
Jour
nal of the Opera
tional
Research
Society
, Vol/Issu
e: 62(12)
. Pp. 20
83-2096, 2011
.
[25]
Bhas
karan, K.
, et al
. “
D
y
n
am
ic
Anycas
t Routi
ng a
nd Wavelength Assignmen
t in
WDM Networks Using Ant
Colon
y
Optim
iz
ation
(ACO)”,
in
2011 Ieee International Con
f
erence on
Communications,
2011.
[26]
Shi, B.
, et
al
. “
A
H
y
brid G
e
net
i
c
&Ant-colon
y Al
gorithm
for Fuzz
y Pe
tri Net
Para
m
e
ter Optim
iz
ati
on Problem
s”, i
n
Mechanical, Ind
u
strial, and
Man
u
facturing
Engin
eering
, M. M
a
,
Editor
.
Pp. 565-
568, 2011
.
[27]
Benbouzid-Sitayeb, F.,
et
al. “An Integrated ACO Approach for th
e Joint Production and Prev
entive Main
tenan
c
e
Scheduling Problem in the Fl
owshop Sequencing
Problem”,
2008 Ieee Intern
ational Symposi
u
m on Indu
strial
Electronics
, Vol. 1-5. Pp. 1541-1
546, 2008
.
[28]
Chandrasekar
an, C., et al. “Metaheuristics for sol
v
ing
economic lot scheduling pr
oblem
s (ELSP)
using time-var
y
ing
lot-siz
e
s appro
a
c
h
”,
European
Jo
urnal of Industrial Eng
i
neering
,
Vol/Issue: 1(2)
.
Pp. 152-181, 20
07.
[29]
Li, A.
and F. B
a
i. “The R-
T
SP model and
its
application in
the VLSI floo
r
plan”,
in
Proceedings of the Fifth
International Co
nference on Info
rmation and Management Scien
ces
,
H.T.
Nguy
en,
X.
Zhao,
and J.
Peng,
Editors
.
Pp. 585-587, 20
06.
Evaluation Warning : The document was created with Spire.PDF for Python.